I. Introduction to the CISG

II. The Doctrine of Good Faith in National Jurisdictions

III. The Doctrine of Good Faith in the CISG

IV. The Application and Enforcement of Article 7

V. Reform

VI. Conclusion

INTRODUCTION

A. The Harmonisation Movement

The last fifty years have witnessed a resurgence in the drive to harmonise international trade law. The work of many international organisations has produced a plethora of legal instruments covering a diverse range of topics. The harmonisation movement stems from the realisation that the absence of a supranational law is inimical to the growth of international trade.

Without a uniform law, any international transaction is potentially governed by a number of domestic laws. The governing law can only be determined by recourse to the forum's conflicts rules. Unless these rules have previously been unified, their content differs from forum to forum. Thus, it is perfectly possible for a plaintiff to "forum shop" for a favourable law. The inability to pinpoint the legal significance of the parties' agreement from its inception undermines the certainty and stability of the transaction.

A choice of law will not necessarily ameliorate all these difficulties. Quite apart from the fact that the forum may not necessarily recognise the parties' choice of law, at least one of the parties may be loath to commit to an unfamiliar law, especially if it is perceived to favour the other. In addition, that jurisdiction's laws may not be sufficiently attuned to the demands of international trade.[1] When a dispute erupts, litigation based on an unfamiliar law (and sometimes in unfamiliar territory) only adds to the costs of the transaction.

A unified law, however, can solve many of these problems. A unification instrument can seek to harmonise either procedural or substantive law. If properly done, a harmonisation instrument will produce one regime that is neutral, clear and comprehensible, thus adding significantly to the stability, predictability and convenience of the transaction. At the same time, a harmonisation effort offers an opportunity to make a clean break from the "awesome relics from the dead past"[2] and institute significant reforms.

B. The Thesis and Organisation of the Paper

An inept harmonisation effort, however, would only compound the problem. A harmonisation effort can be thwarted at two stages - the drafting stage and the application and enforcement stage. This paper seeks to examine these general problems through the lens of one particular problem encountered in the 1980 Vienna Convention on Contracts for the International Sale of Goods ("CISG").[3] The position of good faith in the Convention has proven to be one of the most enduring disputes encountered in the drafting, interpretation and enforcement of the Convention. The different positions that the doctrine [4] occupies in the domestic laws of the various legal families are symptomatic of the gulf in legal traditions that the CISG has had to bridge.

This paper will look at the scope and application of the doctrine of good faith in the CISG in five main parts. The first part will introduce the reader to the birth and the scope of the CISG in general. The remaining parts of the paper focus on the doctrine of good faith itself. In order to understand why the concept of good faith was such a contentious topic during the drafting of the Convention, it is essential to have some sense of the number of competing national approaches to the doctrine. To this end, the second part of the paper offers the reader a general introduction to the doctrine of good faith before turning to the different ways civil law and common law jurisdictions have approached the doctrine. The third part of the paper concentrates on the scope of the doctrine of good faith in the Convention itself. It will examine a variety of possible interpretations and argue that the role of good faith should be limited in the particular circumstances of the Convention. The fourth part of the paper focuses on the problems that have been encountered in the actual application of the doctrine in the context of the CISG. It will demonstrate that judicial and arbitral interpretation and application of the doctrine have only exacerbated problems and discuss the reasons for this phenomenon. In its fifth part, the paper will proffer some suggestions for reform.

I. INTRODUCTION TO THE CISG

A. The Genesis of the CISG

The CISG had its genesis in a suggestion in 1928 by Ernst Rabel that the International Institute for the Unification of Private Law (UNIDROIT) work on unifying the law relating to the international sale of goods. This resulted in two draft Conventions: one relating to a Uniform Law of International Sale and one relating to a Uniform Law on Formation of Contracts for the International Sale of Goods (the "Hague Conventions"). Although these Conventions were adopted at a 1964 Diplomatic Conference at the Hague, they did not meet with widespread success. They were implemented by only 9 States.[5] The instruments were perceived to be the product of continental Western Europe and not representative of the world's legal traditions. In fact, at the 1964 Conference, Latin America was represented only by Colombia, Asia by Japan and Africa by Egypt.[6]

Thus, when the United Nations Commission on International Trade Law (UNCITRAL) resumed work in the area in 1968, it was careful to ensure widespread participation and representation. It allocated permanent seats among the regions of the world and chose its drafting committee in a way that would "represent and insure the involvement, feedback, and support from the distinct legal systems of this world, including the common law, the civil law, the law of communist countries with an affinity to the civil law but with a different ideological basis, the Islamic law, Hindu law, Chinese law, and numerous systems with mixtures thereof."[7]

Ironically, its strength was also its weakness. The diverse representation multiplied the number of potential fault lines and increased the scope for disagreement. Indeed, the delegates often found themselves split according to legal tradition, economic system and stage of socio-economic development.[8] The final draft is an interesting mix of hard-won compromises.[9]

Nevertheless, the final draft of the Convention was adopted in 1980 at a Diplomatic Conference in Vienna attended by 62 countries and 8 international organisations. The Convention entered into force on 1 January 1988 upon the ratification of 11 States. Since then, the CISG has been adopted by 61 States [10] and has been triumphantly heralded as the "uniform international sales law of countries that account for over two-thirds of all world trade.account for two-thirds of all world trade".[11]

B. The Scope of the CISG

The CISG is divided into four parts. Part 1 lays down the Convention's sphere of application and general provisions. Part 2 deals with the formation of the sales contract while Part 3 contains provisions relating to the substantive obligations of both the buyer and the seller.[12] Part 4 deals with the public international law aspects of the Convention.

Significant omissions from the Convention are provisions relating to the validity of the contract,[13] the effect of the contract on the property in the goods sold,[14] and the liability of the seller for death or personal injury caused by the goods.[15]

According to Article 1, the Convention applies to contracts of sale of goods between parties whose places of business are in different States when (a) those States are Contracting States or (b) when the rules of private international law lead to the application of the law of a Contracting State.[16]

The CISG, however, does not deal with those contracts that are enumerated in Articles 2 and 3 of the Convention.

The parties are free to exclude the application of the Convention or vary or derogate from the effect of any of its provisions.[17]

C. The Interpretation of the CISG

The CISG does not lay down any specific methodology to be employed in interpreting the Convention. Guidance, however, may be derived from two main sources in the Convention- the preamble and Article 7.

The preamble lays out the purposes of the Convention: inter alia, to promote a New International Economic Order and to encourage "friendly relations among States" by developing international trade "on the basis of equality and mutual benefit". In terms of practical utility however, the precatory and aspirational terms of the preamble do not provide much concrete assistance.[18]

Article 7, on the other hand, is more helpful. Due to the central role that this provision plays in the debate over the role of good faith in the CISG, it is useful to reproduce the provision here:

This provision clearly requires arbiters and judicial officers to treat the Convention as a creature autonomous from domestic laws. The Convention sets out a hierarchy of sources for the resolution of a dispute. Regard is to be had in the first instance to the express provisions of the Convention. In the event that they do not provide a solution, the adjudicator should refer to the general principles on which the CISG is based. National laws should be used only as a last resort. Indeed, in interpreting the Convention, decision-makers must not be quick to "read in" domestic concepts.

Article 7, however, also stands at the heart of a controversy over the proper role of good faith in the CISG. Article 7 is the only place in which the concept of good faith is expressly referred to in the CISG. Unfortunately, the wording of Article 7 reveals some internal contradiction. Ostensibly, Article 7(1) of the Convention allocates only an interpretative role to the doctrine. Article 7(2) of the Convention, however, permits the use of the general principles underlying the CISG to resolve matters not expressly settled by the instrument. It is certainly arguable that the principle of good faith underlies many of the provisions in the Convention.[20] Article 7(2), therefore, could well accord the doctrine of good faith a far more substantive role in the Convention.

This structure has been variously termed as "uneasy"[21] and "strange".[22] As Eörsi put it: "it was widely thought that the rule was vague, or at least would remain vague for a long time, and, because of the laconic language of the CISG, would never become unambiguous."[23]

II. THE DOCTRINE OF GOOD FAITH IN NATIONAL JURISDICTIONS

To put it mildly then, Article 7 is awkwardly drafted. To better appreciate the cauldron in which this provision was forged, it is apposite to first examine the competing national approaches to the doctrine of good faith. The debate over the role of good faith in the CISG can be examined along several divisions. The debate over this issue raged not only between the common law and civil law delegations, but also between the socialist and capitalist representatives and even among representatives who shared a common cultural and legal background.[24] While the Western representatives argued, for example, that a substantive good faith obligation was too vague, the socialist representatives contended that it was both practical and flexible.[25]

It is proposed, however, to examine the doctrine along common law-civil law lines since the debate was mostly between these two camps. Two caveats, however, must first be entered. Firstly, this paper will only be examining the doctrine of good faith in the context of contracts for the sale of goods. The doctrine may well have a different role to play in other areas of the law.[26] Secondly, this part of the paper is merely meant to provide a brief flavour of the role that good faith plays in a number of national jurisdictions. The proper ambit of the doctrine in the national sphere is itself the subject of much debate and a reader who is interested in this area should refer to the burgeoning jurisprudence in this area.

A. A General Introduction to the Concept of Good Faith

The concept of good faith [27] has great normative appeal. It would be a rare commentator who would deny that good faith has any role to play in the development of a legal system. After all, it is the aspiration of every mature legal system to be able "to do justice and do it according to law".[28] A distinction, however, must be made between good faith as a mere ethical or moral principle that every legal system aspires to and good faith as a legal doctrine that is capable of independently generating concrete legal rights and obligations on its own. This paper focuses on whether it is viable to think of "good faith" in the latter sense.

There is no doubt that there are many advantages associated with a legal doctrine of good faith. From a substantive point of view, a doctrine of good faith is the antithesis of formalism and positivism. A doctrine of good faith could safeguard the expectations of contracting parties by respecting and promoting the spirit of their agreement instead of insisting upon the observance of the literal wording of the contract. It could also regulate self-interested dealings.

At the same time, a doctrine of good faith could reduce costs and promote economic efficiency. Parties need not assiduously contract for every contingency because the doctrine of good faith can be relied upon to fill in unforeseen gaps. From a methodological point of view, a doctrine of good faith is a far more rational approach to the resolution of issues of bad faith than a piecemeal approach. A doctrine of good faith would provide a sound theoretical basis to unite and undergird what would otherwise appear to be merely a series of disparate rights and obligations.

The above hypothesis, however, is based on the assumption that the doctrine of good faith is coherent and well developed. The danger of a doctrine of good faith is that it can lapse into generality and vagueness. It can threaten to become so unpredictable that it increases the costs of transacting and undermines the legitimate expectations of contracting parties.[29] As Professor Goode puts it, "one may acknowledge the power and attraction of a general idea but the idea may be so general that it is of no practical utility to the merchant".[30]

In addition, the adoption of such a doctrine represents a significant departure from the classical model of contract and the idea of freedom of contract. Courts and arbiters are now required to take on a more activist role and impose substantive obligations that may, at times, even contradict the express wording of the contract.[31] In an extreme case, judges may even be tempted to impose their own moral conceptions on the parties.[32]

Unfortunately, there is no definite way to resolve this tension between justice and certainty. Broadly speaking, problems lie with two aspects of the doctrine - its definition and its form. A coherent definition is first necessary to form a stable base from which adjudicators can elucidate a series of good faith rights and obligations. Even then, however, a definition would not necessarily dictate the form that the doctrine must eventually take. There are multiple directions that can be taken in the process of translating the general to the concrete.

Unfortunately, there is no single global definition of good faith. The doctrine of good faith in domestic sales law is a slippery concept with multiple connotations and multiple meanings. At most, we can say that the definition of "good faith" can be framed either positively or negatively. In other words, the content of "good faith" can be elucidated either by reference to what it is or by reference to what it is not.

A "negative" definition of good faith is probably the easiest to describe. As Professor Summers explains, "in cases of doubt, a lawyer will determine more accurately what the judge means by using the term 'good faith' if he does not ask what good faith itself means, but rather asks: What…does the judge intend to rule out by his use of this phrase?"[33] In other words, one "homes in" to a definition of good faith by eliminating what would be considered clear instances of bad faith rather than by attempting to determine what would be considered instances of good faith. The drawback of this approach, however, "is that it seems tantamount to saying that the good faith duty is breached whenever a judge decides it has been breached…[which] hardly advances the cause of intellectual inquiry and…provides absolutely no guide to the disposition of future cases, except to the extent that they may be on all fours with a decided case".[34]

In so far as positive definitions are concerned, there are many. "Good faith" has been variously defined as "fairness",[35] "fair conduct",[36] "reasonableness",[37] "reasonable standards of fair dealing",[38] "good faith and fair dealing",[39] "community standards of decency, fairness or reasonableness",[40] "honesty in fact",[41] "decent behaviour",[42] "a common ethical sense"[43] and as "a spirit of solidarity".[44]

At best, these definitions merely replace the concept of good faith with equally vague and nebulous terms. They fail to pinpoint its meaning. Worse, many commentators use the terms interchangeably, insensible to the subtle nuances between one term and the next.[45] Take the concepts of "honesty", "fair dealing" and "reasonableness". It is true that all three seek to moderate unethical conduct. However, they actually refer to very different things. "Honesty", for example, refers to a subjective state of mind[46] while "fair dealing" refers to an objective state of affairs. Furthermore, one can conceivably act unreasonably without being dishonest.

At the end of the day, the meaning of "good faith" seems to be predicated on one's intuitive sense of justice. The danger of this approach, however, is that it may encourage a tendency to equate everything that appears to promote justice with good faith. The meaning of "good faith" becomes so general and abstract that it is meaningless. This is hardly a strong foundation for a "global" doctrine of good faith in an international sales convention.

Matters are further complicated by questions concerning the issue of form. Issues that would have to be resolved include:

B. The Competing National Approaches

Let us now turn to look more specifically at the doctrine as it has developed in individual jurisdictions. The principal jurisdictions that will be examined are the common law jurisdictions of the United States of America, United Kingdom, Australia and Canada and the civil law jurisdictions of Germany and France.[50] The differences in these jurisdictions' approach to good faith can be divided into three categories: differences between legal families, differences among legal systems within the same legal family and differences within a single legal system. These categories will be examined in turn.

1. Differences between legal families- the Civil Law vs. Common Law

The doctrine of good faith has a much older lineage in the civil law than in the common law. This probably stems from "the influence of Canon law on the Civil law and of commercial practice on the Common law. A consequence of this commercial orientation of the Common law is that if a choice has to be made between certainty and justice in the individual case, it is likely to be made in favour of certainty."[51] The civil law, on the other hand, is "primarily concerned, not with the economic exchange between the parties, but with the exchange of consents and with the moral evaluation of the behaviour of the parties."[52]

In the civil law, the doctrine of good faith typically takes the form of a general codal provision.[53] One of the most famous examples is § 242 of the German Civil Code ("BGB") which provides that the obligor is bound to perform the contract in good faith (Treu und Glauben mit Rücksicht auf die Verkehrssitte), having regard to common usage.[54] Broadly speaking, the German doctrine of good faith has three basic functions: "it serves as the legal basis of interstitial law-making by the judiciary, it forms the basis of defences in private law suits, and it provides a statutory basis for reallocating risks in private contracts".[55]

To a common law lawyer, the scope of § 242 is overwhelming. § 242 of the BGB is the progenitor of such an elaborate structure of rules that over 500 pages of the 13th edition of the Staudinger commentary is devoted to it.[56] The requirement of good faith has been applied in such a wide variety of situations that the term "good faith" is not susceptible to a single definition. There appears to be no single principle that unifies the entire range of cases in which the provision has been used. Thus, commentators are generally forced to take a piecemeal approach and explain § 242 by reference to the various categories of situations in which it has been applied.[57]

Two examples will illustrate the scope § 242 has provided for judicial creativity.[58]

According to the strict letter of the BGB, a creditor can claim damages for what in common law parlance would be a "breach of contract" in only three categories: if performance has become impossible for the debtor (Umnoglichkeit), if the debtor's fault has caused a delay in performance (Schuldnerverzug) or if the seller fraudulently concealed the defect in defective goods or warranted a promised performance (Gewashrleistung).[59] The German courts have thus turned to § 242 of the BGB for help in creating a residual cause of action for breaches not falling within these three categories.[60]

Judicial creativity has also been essential in the creation of the doctrines of frustration and impracticability.[61] For example, "[w]hen the German currency collapsed following the First World War and the mark plummeted to less than a billionth of its value at the time of the armistice, it was the obligation of good faith that was used as the principal basis of revaluation of debts in terms of the real purchasing power of the mark in spite of specific statutory provision that the mark was legal tender."[62]

In comparison, with the exception of the United States, common law jurisdictions generally do not have a doctrine of good faith.[63] Instead, common law jurisdictions adopt "piecemeal solutions in response to demonstrated problems of unfairness".[64] If we construe the concept of good faith broadly as simply a mechanism which is used to moderate the effects of unethical behaviour or ensure justice and fairness by requiring contracting parties to observe certain behavioural norms, we can say that the idea manifests itself in specific doctrines, such as duress, undue influence, mistake and misrepresentation; or in specific statutes such as the English Sale of Goods Act [65] and the Unfair Contract Terms Act. Outside such doctrines, parties are free to pursue their strict legal rights regardless of their motive.[66]

While the doctrine of good faith in the civil law does bear a certain affinity to the common law doctrine of implied terms, it is generally much harder to imply a term in the common law than it is to find a good faith obligation under a general doctrine. In English law, for example, a term is generally implied only on the basis of a "business efficacy"[67] or "officious bystander"[68] test. A term will not be implied simply on the basis that it is reasonable or fair to do so.[69]

While the common law may often reach the same result as the civil law does through the judicial development of general principles,[70] the fact remains that the existence of a general doctrine provides decision-makers with a more malleable and flexible standard. A general doctrine stands at the ready to meet any unforeseen contingencies. As was alluded to above, § 242 of the German Civil Code has been used to deal with legislative oversights or omissions in the Code, clarify concepts, create and implement new obligations on the parties and as a basis for new remedies.[71]

While case technique in the common law does operate on the identification of the general principles underlying specific fact situations, it has yet to progress so far as to identify principles that overarch entire areas of the law.[72] The general principle, if any, that is extracted from the cases remains unique to the specific doctrine. For example, at the risk of over-simplification, the common law may find a general principle, based on a series of cases, that a defendant is guilty of fraudulent misrepresentation if he makes a false statement of fact knowingly, without any belief in its truth, or recklessly.[73] The common law may find another principle based on another series of cases that a contract is void on the basis of mistaken identity when the plaintiff intended to contract with and only with, the person the defendant misrepresented himself to be.[74] The common law, however, will not go one step further and derive a general principle based on these cases that parties have a general obligation in contract law to perform and enforce their rights in good faith.

The other major difference between common law jurisdictions and civil law jurisdictions lies in their treatment of pre-contractual acts of bad faith. The civil law generally has a well-developed doctrine of pre-contractual liability.[75] The German doctrine of culpa in contrahendo, for example, holds a defendant liable for damages when he "awakes in the other confidence in the imminent coming into existence of a contract- subsequently not concluded- and thus causes the latter party to incur expenses".[76]

Most common law jurisdictions do not have a regime of pre-contractual liability of bad faith.[77] The common law generally takes an aleatory view of contractual negotiations and accords parties the freedom to break off negotiations with impunity.[78] This is true even of the United States, which has the most well-developed general doctrine of good faith among the common law jurisdictions. The United States is forced to arrive at the same results through the use of specific doctrines such as unjust enrichment, misrepresentation and specific promise.[79] Unlike the civil law, however, the United States would not impose liability on a party who refuses to contract without an appropriate ground even when he has "conducted himself in such a fashion that the other party could, and did justifiably count on a contract with the negotiated content coming into existence".[80]

2. Differences among legal systems in the same legal family

a. Differences among common law jurisdictions - the United Kingdom (no general doctrine of good faith) vs. Australia and Canada (emerging) vs. the United States (developed)

The above sub-section has perhaps over-simplified the issue by understating the differences among legal systems in the same legal family and painting the debate as a battle between polar opposites. The next two sub-sections attempt to redress this imbalance by focusing on the differences among and within legal systems in the same legal family.

The doctrine of good faith is in varying states of growth in common law jurisdictions. The United Kingdom and the United States stand at opposite ends of the spectrum. A general doctrine of good faith is non-existent in the United Kingdom. Instead, it has developed piecemeal solutions in response to specific problems of unfairness. As the English Court of Appeal put it, "in the case of commercial contracts, broad concepts of honesty and fair dealing, however laudable, are a somewhat uncertain guide when determining the existence or otherwise of an obligation which may arise even in the absence of any dishonest or unfair intent."[81]

One reason for what some may consider an uncompromising approach is the country's emphasis on promoting certainty and predictability in the world's leading financial centre. As Professor Goode puts it:

It must be acknowledged, however, that there has been an upsurge in interest in the doctrine among English commentators in recent years.[83] This has been especially fuelled by the introduction of the 1993 EC Directive on Unfair Terms in Consumer Contracts, which makes several references to the concept of good faith.[84] There is little sign, however, of any judicial development of a general doctrine of good faith.[85]

On the other hand, we have the United States, which does have a developed doctrine of good faith. The doctrine has been used to proscribe, inter alia, attempts to evade the spirit of the deal, wilfully render an imperfect performance, abuse a discretion conferred by a contract, interfere with the other party's performance or refuse co-operation.[86]

The doctrine of good faith has been recognised in two major places - the Uniform Commercial Code ("the UCC") and the Restatement (Second) of Contracts. § 1-203 of the UCC specifically requires the observance of good faith in the performance or enforcement of every contract or duty within the Act.[87] "Good faith" is defined in § 1-201(19) as "honesty in fact in the conduct or transaction concerned", unless the context otherwise requires. "Good faith" in the case of a merchant, however, is defined by § 2-103(1)(b) to mean "honesty in fact and the observance of reasonable commercial standards of fair dealing in the trade".

On the other hand, while § 205 of the Restatement also imposes upon each contracting party a duty of good faith and fair dealing in the performance and enforcement of the agreement, it does not adopt a positive definition. Rather, the Restatement chooses to define the concept of "good faith and fair dealing" by vague reference to its purposes and admits that "good faith" can only be recognised by what it is not:

In other words, this definition adopts what Professor Summers call the "excluder analysis"- the concept of "good faith" is defined by reference to what it is not, rather than what it is.[90]

Compared to the United States, Australia and Canada occupy intermediate positions. Due to historical reasons, these legal systems bear close affinity to the English system. Since these jurisdictions abolished the right to appeal to the English Privy Council, however, they have been free to develop their legal systems in an autochthonous manner.

The doctrine of good faith is still in its infancy in Australia. Judicial sentiment appears divided over the question of the viability of a general doctrine of good faith.[91] Chief Justice Sir Anthony Mason [92] and Judge Priestley [93] have come out in favour of the doctrine in both judicial and academic writings. Justice Gummow, however, has expressed scepticism about the ability of the doctrine to provide any more guidance than established principles of contract law already do.[94]

In Canada, on the other hand, the debate is not so much over whether to adopt a general doctrine of good faith, but which standard to adopt.[95]

b. Differences among civil law jurisdictions

Even though the doctrine of good faith is commonly associated with the civil law, not all civil law jurisdictions have embraced it as enthusiastically as Germany. France is a significant example of restraint. Like Germany, France too has a general codal provision that mandates the observance of good faith (Article 1134 of the French Civil Code).[96] Unlike Germany, however, this provision has been used purely in an ancillary fashion. As Professor Bridge puts it, "[g]ood faith is enshrined in art. 1134 of the French civil code but its practical impact can be described as shallow: it has done nothing to disallow penalty clauses, it has not expanded the narrow categories of lesion and it has not been employed to give relief in what we would now call cases of commercial impossibility".[97] It is thought that French judges are perhaps as concerned as their English brethren are on the impact that a wide-ranging doctrine of good faith would have on the certainty of legal relations.[98]

3. Controversy within a legal system

The controversy over the appropriate scope and application of good faith does not cease at national borders. The meaning of good faith has proven to be so malleable and elusive that there is disagreement over the ambit of the doctrine even within those jurisdictions that have adopted it or that are favourably inclined towards it. This appears to be more true of common law jurisdictions than civilian ones. The two common law examples that will be used here are the United States and Canada.

Although an obligation of good faith has been legislatively imposed in the United States by the Uniform Commercial Code, there are difficulties at both a theoretical and a practical level.[99] At the theoretical level, dissatisfaction exists over the appropriate scope and application of the concept of good faith. At least three main lines of thought have developed.

Professor Farnsworth is of the view that the significance of the doctrine is in "implying terms in the contract".[100] Professor Summers is of the opinion that no clear meaning can be ascribed to the doctrine of "good faith". At best, it is used to rule out or exclude specific instances of bad faith (the "excluder analysis").[101] Professor Burton, agrees that "neither courts nor commentators have articulated an operational standard that distinguishes good faith performance from bad faith performance."[102] He points out, however, that a contract will normally confer on another discretion over some aspect of contractual performance, such as price, output, quantity, or time of performance. He argues that the concept of good faith "limits the exercise of discretion in performance conferred on one party by the contract."[103] It is bad faith to use one's discretion "to recapture opportunities forgone in contracting"[104] as determined by the other party's reasonable expectations. The judiciary has come out in support of all three standards, sometimes without even recognising the differences between them.[105]

To make matters worse, it is not entirely clear how the academic debate in this area squares with the competing definitions in the UCC and the Restatement (Second) of Contracts. On one theory, each commentator is merely interpreting the prevailing standards of good faith. On another, he is in fact proposing an independent version.

Even if the debate over the definition of "good faith" could be resolved, there is still controversy over whether the concept of good faith should be measured against an objective or subjective standard.[106]

At a more specific level, there is some doubt as to as to the practical utility of the doctrine. Some commentators are of the view that the doctrine does not achieve any more than what established principles of contract law already accomplish. As Professor Farnsworth admits, "many of the uses to which the new concept of good faith is put today do not go beyond those to which the traditional techniques of interpretation and gap filling were traditionally put".[107]

In addition, even where cases are ostensibly decided under the good faith provisions of the UCC, the results are not necessarily consonant with the definition of good faith in the Act. In fact, Professor Burton is of the opinion that the analysis of the courts is ultimately more consonant with the "foregone opportunities" mode of analysis rather than the standards of "honesty" and "reasonable commercial standards of fair dealing" that have been expressly provided for.[108]

In Canada, there is a similar debate over the appropriate standard to adopt:

Cases can be found that support either standard. The significance of the distinction lies in the fact that the "reasonableness standard" can be used to override the express terms of the contract, while the "interpretative standard" would only impose obligations of good faith which are consistent with the parties' contractual expectations.

Two cases illustrate the contrasting results that these two approaches produce. In Watchfield Developments Inc v Oxford Elgin Developments Ltd,[110] the court used the reasonableness standard in a way that negated the utility of clauses that provide that "time is of the essence". It held that the purchaser could rely on such a clause only if he could show that he would have completed but for the vendor's delay. This could be proved only if the purchaser had "agreed to a short extension of time for closing".[111] In Khangura v. Triple R. Construction Ltd,[112] on the other hand, the purchaser was entitled to rely on a "time is of the essence" clause to sue for the return of a deposit. The court applied the "interpretative standard" and saw no reason why the purchaser could not rely on what it had expressly bargained for. The risks could have been foreseen and it was up to each party to protect its own position.

Needless to say, the debate over the appropriate standard to adopt within United States and Canada inhibits the development of a clear and coherent doctrine.

III. THE DOCTRINE OF GOOD FAITH IN THE CISG

With all this doctrinal untidiness among and within the legal families, the drafting committee and the working groups of the CISG certainly had their work cut out for them. The task was further complicated by the fact that even if it were the case that the meaning of good faith was settled in the domestic context, the demands of good faith in the international context were different. The warning sounded by the UNIDROIT Principles of International Commercial Contracts is equally applicable here:

The paper will now examine the appropriate role of good faith in the Convention. A brief word, however, must first be said on the appropriate technique to be used in interpreting an international convention. Recourse must be had, of course, in the first instance, to the wording of the provision and its context. Assistance, however, may also be sought from the travaux préparatoires. In so far as the CISG is concerned, the travaux préparatoires of the Convention are remarkably accessible.[114]

According to Article 32 of the Vienna Convention on the Law of Treaties, the preparatory work of the Convention and the circumstances of its conclusion may be used to confirm or determine the meaning of a provision.[115] A number of major jurisdictions observe this practice.[116] Indeed, Article 32 can be said to codify a practice of customary international law.[117]

Of course, one must exercise caution when utilising the travaux préparatoires. The delegates may not necessarily have dealt with the problem at hand; or if they did, the indications may be so conflicting that the negotiating history of the Convention is of no help at all. Moreover, "[o]nce adopted, the Convention, like any other law, has a life of its own, and its meaning can change with time so that the intention of the parties is only one of the elements to be taken into account for the purpose of its interpretation".[118] However, the travaux préparatoires cannot be ignored when they clearly demonstrate that one interpretation was intended.

It is thus appropriate to make a brief excursion into the history of Article 7. An examination of the legislative history reveals that every effort to introduce a substantive obligation of good faith and fair dealing was repelled.

A. The Drafting History of Article 7

1. The first foray into the field of good faith [119]

The introduction of good faith into the Convention was first suggested by a representative of Spain in 1972.[120] It was not until 1978, however, that the topic of good faith was revived and subjected to sustained debate. At the eighth session [121] of the Working Group, the following provision was tabled for discussion:

In so far as the first paragraph was concerned, "the general concept that the draft Convention should contain provisions relating to good faith and fair dealing was supported by a majority of the representatives".[123] It was felt that similar principles had been useful in many jurisdictions and that they had the potential to serve equally well in the international context. Although uniformity in interpretation may be difficult to achieve initially, this problem would not be any worse than the situation that first existed in national jurisdictions after they adopted general provisions requiring the observance of good faith.

However, there was already some concern over the specifics of the provision at this time. The first major concern was that the provision was too vague and imprecise, with the danger that its meaning would depend on subjective value judgements. The second concern was that the provision would not be useful until it had been subjected to a sustained period of judicial interpretation.

The remaining two paragraphs failed to garner much support, in part because of their vagueness. It was also felt that the issue of exemption clauses was not appropriate in a convention dealing with transactions between merchants, "where exclusions of liability for the seller were frequently compensated by a lower price for the buyer".[124]

The Working Group thus decided to abandon the last two paragraphs of the provision and adopt an amended version of the first (which was later renumbered Article 5):

2. Good faith in retreat [126]

This provision, however, was the subject of a fierce counterattack at the eleventh session of the Commission in 1978. A considerable number favoured the deletion of the provision on the grounds that an international concept of good faith was too vague to be of any practical utility. Decision-makers would necessarily have to fall back on domestic notions of good faith, with the result that different interpretations would be given to the provision in different countries. The resulting uncertainty would be detrimental to international trade. Another view was that it was unnecessary to include a requirement of good faith because it was already implicit in all laws regulating business activity.[127]

The provision was also attacked on the basis that it did not specify any sanctions for its breach. It was considered ironical that the UNIDROIT draft text on validity [128] found it necessary to closely regulate the consequences of clear breaches of good faith while it was assumed that the consequences for less serious breaches of good faith would be obvious to adjudicators interpreting the Convention.[129] It ignored the fact that adjudicators would seek assistance from their domestic training at the cost of uniformity. Thus, it was considered that the most appropriate place for a provision on good faith and fair dealing was in a Convention which dealt with the validity of contracts.

There was, however, also a considerable number who supported the retention of the article. It was felt that the concept of good faith was universally recognised. Although the concept of good faith was not necessarily the same in every jurisdiction, its content would be elucidated incrementally through judicial decisions. Although the provision did not specify any sanctions for its breach, this could be determined by domestic courts in a flexible manner in response to the individual facts of the case. Moreover, "even without sanctions the existence of the provision would draw the attention of the parties and the court to the fact that high standards of behaviour were expected in international trade transactions".[130] Further, it was hoped that the adoption of the provision would promote some of the aspirations of the proposed "new international economic order" and lessen discriminatory or undesirable trade practices. It was feared that the deletion of the provision would expose the Commission to the charge that it believed that good faith was unnecessary in international trade.

A number of those who otherwise supported the article, however, took exception to the specific formulation of the article. Many of the developing countries feared that "fair dealing" would be taken to refer to current standards of international business practices, when these standards were not necessarily "fair" to developing countries. Consequently, they preferred to have the reference to fair dealing deleted than risk the elevation of these practices into rules of law. A proposal that the concept of "fair dealing" be replaced by one of "international co-operation" only compounded the problem. The term was so vague and nebulous that its precise scope and effect could not be determined.

As a way out of the impasse, a Working Group was established to seek a compromise. It considered a number of compromise solutions. The first suggestion that the concept of good faith be incorporated into the preamble was rejected on the basis that it would render it devoid of any effect. Another suggestion was to incorporate the concept into the provision on the interpretation of the statements and conduct of the parties. This too was rejected on the basis that the purpose of Article 5 was not to determine the intent of the parties but to impose a standard of behaviour.

The most widely supported compromise was a newly worded Article 6 (now Article 7(1) which read:

"In the interpretation and application of this Convention regard is to be had to its international character and to the need to promote uniformity and the observance of good faith in international trade."[131]

3. The death of good faith

a. Death of the attempts to include a more pointed reference to good faith [132]

At the 1980 Vienna Conference, repeated attempts to resurrect the concept of good faith failed. Norway and Italy objected to the confinement of the concept of good faith to the interpretation of the Convention. Norway suggested that the reference to good faith in Article 6 be deleted and be transferred to Article 7 (now Article 8) to "make it clear that the principle of good faith was relevant to the interpretation of the contract of sale, but not to the interpretation of the future convention as such".[133] Italy's suggestion was more extensive. Its proposal entailed the deletion of the reference to good faith in Article 6 and the adoption of a new article that would make good faith relevant to the formation, interpretation and performance of the contract.[134] This proposal was overwhelmingly rejected in favour of Article 7(1) as presently worded.[135] The general sentiment was that the provision as presently worded was already the result of a hard-won compromise and there was little reason to re-open the issue again.

b. Death of the attempts to prohibit the parties from departing from the obligation of good faith and to provide for pre-contractual liability

At the 1980 Conference, two other proposals involving the concept of good faith were made. The first was a proposal by Canada that parties be prohibited from excluding the obligations of good faith, diligence and reasonable care unless they wholly excluded the Convention.[136] The second was a proposal by the German Democratic Republic that the Convention provide for pre-contractual liability.[137] These proposals were rejected for a variety of reasons, not all of which had to do with objections against the concept of good faith. One of the objections to these proposals, however, was that they represented an attempt to sneak an obligation of good faith through the back door.[138]

B. The Role of Good Faith in the Convention

1. The Competing Interpretations of the Role of Good Faith in the Convention

In summary, then, all the attempts to introduce substantive obligations of good faith (or substantive obligations related to the concept of good faith) were rejected. Despite the clarity of this point, an overwhelming number of commentators are of the opinion that the Convention imposes a substantive obligation of good faith on contracting parties.[139]

The remainder of this part of the paper will examine the viability of this view. I will first identify and explain the number of competing answers to the issue before identifying the most appropriate one for the Convention. In examining the meaning of good faith in the CISG, six different positions can be identified:

a. An obligation of good faith established by the practices of parties or by international trade usages.

The debate over the place of good faith normally centres on the appropriate scope of Article 7. However, during the debates on good faith, some of the delegates had argued that it was unnecessary to include an explicit reference to the concept because it was already an implicit requirement of all business activity. The basis of their argument could have been Article 9 of the CISG. Article 9 states:

Article 9 identifies two sources of rights and obligations: usages and practices.

It is conceivable that parties may establish a practice of good faith or fair dealing among themselves in their course of dealing.

A usage may also build up in the trade requiring the observance of good faith. Even if no trade usage exists which positively requires the observance of good faith, usages may still help to indicate what conduct would be considered unacceptable. The latter argument bears some resemblance to Professor Summer's thesis that while it may not be possible to define with certainty what "good faith" means, it is at least possible to determine what it is not. However, as pointed out above, there are serious problems associated with defining the concept of good faith negatively.

b. The doctrine of good faith as an interpretative guide.

This interpretation takes Article 7 of the CISG on its literal wording. The concept of good faith is expressly mentioned only in Article 7(1) of CISG and this provision accords the concept of good faith merely an interpretative role. Thus, the doctrine of good faith should only be used to interpret the Convention.

c. The doctrine of good faith as a prerequisite to the exercise of the rights and remedies provided in the Convention.[141]

This interpretation is based on Article 7(2) of the CISG whereby "good faith" is one of the "general principles" on which the CISG is based. Under this interpretation, the obligation of good faith only qualifies that which has been expressly provided for in the Convention. In other words, the buyer or the seller must act in good faith before he can exercise any of the rights or remedies that have been expressly provided for by the CISG.

For example, a party may not be allowed to compel specific performance under articles 46 (for the buyer) or 62 (for the seller) or avoid the contract under Articles 49 (for the buyer) or 64 (for the seller) after a market change if these remedies would allow him to speculate at the other's expense.[142]

d. The doctrine of good faith as a substantive principle in the resolution of matters not expressly governed by the Convention.

This interpretation is also based on Article 7(2) of the Convention. This interpretation, however, goes one step further and accords the concept of good faith an even more substantive role. The concept of good faith now does not merely qualify what has been expressly provided for in the CISG but also serves as an independent general principle to resolve questions concerning matters governed by the Convention but which are not expressly settled in it.

e. The doctrine of good faith as a source of rights and obligations which may contradict or extend those in the Convention.

This is clearly an inappropriate use of the doctrine but it is mentioned here for the sake of completeness.

The problem often arises when concurrent remedies are available in domestic law. The issue then is whether it is permissible to take advantage of these remedies as well as those provided by the CISG. In so far as the CISG does not provide expressly or implicitly for any rights or obligations, the buyer or seller may take advantage of these remedies. Otherwise, the general rule is that:

In other words, the principle of good faith must not be used to contradict or extend what already has been provided for in the Convention. Due to the obvious inappropriateness of this interpretation, its viability will not be discussed any further.

f. The doctrine of good faith merely serves as an interpretative guide. In cases not expressly governed by the Convention, general principles may be derived which are, at best, particular manifestations of good faith. A general doctrine of good faith, however, does not exist to serve as a fount of additional rights and obligations.

Under this interpretation, Article 7(2) does not give rise to a general principle of good faith in a form that is able to generate answers to matters governed by the CISG but which are not expressly settled in it. The provision, however, may give rise to certain general principles that can be described as particular manifestations of a general idea of good faith.

In this context, "good faith" is interpreted broadly to refer to anything that would require contracting parties to behave in a manner that would promote justice, fairness or ethical behaviour. Thus, principles such as the rule against undue influence, mistake and misrepresentation would be considered specific embodiments of this idea of good faith. However, they would not together form the basis for a substantive doctrine of good faith that could independently generate legal rights and duties. At most, the term "good faith" would simply serve as a convenient compendious term for more particular principles or as a label for a moral aspiration.

The only place where good faith could play a role in a legal rather than an ethical sense would be in the interpretation of the Convention. This point is susceptible to at least two variations. One would be that good faith in the legal sense would not have a role to play even in the interpretation of the Convention. Instead, the term "good faith" in Article 7(1) would simply be read to refer to only those principles in the Convention that attempt to achieve justice and ensure ethical behaviour by requiring contracting parties to observe certain behaviour. The other is that the term "good faith" in Article 7(1) would be read to refer to anything that can colloquially be said to promote justice and ethical behaviour, whether it be embodied in the CISG or otherwise. The second variation, however, is impractical as it is too expansive. Its feasibility will thus not be discussed any further.

2. The Appropriate Role of Good Faith in the CISG

Simplifying matters, there are basically four possible roles for the doctrine of good faith in the CISG: as a practice of the parties, as a trade usage, as a substantive general principle or as an interpretative tool.

The appropriateness of each role will be discussed in turn. In determining the most appropriate role for good faith within the Convention, one should be careful not to be beguiled by the normative appeal of the concept of good faith. Although there is no doubt that it would be desirable for the Convention to promote the cause of good faith to the widest extent possible, it should perform this function only if the doctrine of good faith is proved to be a coherent one.

a. Good faith as a practice of the parties

It is submitted that one should not conclude that a general doctrine of good faith exists in the Convention from the fact that some contractual parties may establish 'good faith' practices among themselves. As the word 'practice' connotes, the existence and content of any 'good faith' practice would be dependent on a number of factors, such as the type and importance of the contract, the length and nature of the relationship, and the parties' relative bargaining power. It would be not be advisable to infer that a general regime of good faith rights and obligations exists from a collection of disparate practices.

b. Good faith as a trade usage

To reiterate, Article 9(2) of the CISG requires that the usage, if any, must be one that "the parties knew or ought to have known and which in international trade is widely known to, and regularly observed by, parties to contracts of the type involved in the particular trade concerned".

The strongest argument that can be used here is that an obligation of good faith can be derived from the new lex mercatoria.[145] The lex mercatoria has been defined as an autonomous system of transnational commercial law that is "capable of being applied by decisionmakers (judges or arbitrators) as a source of legal rules, in order to give content to decisions, in much the same way that these decisionmakers would apply a real legal system such as the lex fori or the lex loci arbitri".[146]

Many commentators are of the opinion that "good faith" is a component of the lex mercatoria.[147] The obligation of good faith supposedly implies at least three other duties: "the duty to inform the other party of circumstances which might threaten the performance of the contract; the duty to renegotiate the contract in order to salvage the commercial relationship if circumstances permit; and the duty to mitigate damages in the event of a breach".[148]

Some commentators object to the lex mercatoria on the basis that it is impossible for an anational legal system to exist.[149] They argue that a legal system can derive binding force only from the recognition of a sovereign nation. Putting this objection aside, there are still many problems associated with the argument that good faith is a valid trade usage.

Firstly, the lex mercatoria is comprised not just of uncodified custom or usage but also public international law, uniform laws, the general principles of law, the rules of international organisations, codifications of customs and usages by international organisations, standard-form contracts and reported arbitral awards.[150] Article 9(2), however, only speaks of usages and practices. Unfortunately, when commentators or arbitral awards refer to good faith as a component of the lex mercatoria, it is often not clear how that conclusion was arrived at or which one of the sources were used. The conclusion could well have been based on a source other than trade usage.

The Norosolor case is a fairly typical example of this problem. This arbitral case held that:

The arbitral tribunal's discussion of the existence of good faith as a component of the lex mercatoria does not go any further than this. Its analysis is vague and unhelpful. The reader is left clueless as to the precise theoretical basis for its conclusions. Moreover, its finding that the emphasis on good faith is one shared by all national laws is suspect. As was discussed above, legal families and even legal systems within the same legal family have very different ideas of what it means to act in good faith. Even if were the case that good faith were a principle common to all national laws, it is still different from a usage which requires at the very least, "a practice or pattern of behaviour…established by repetition".[152]

Even if it were the case that the lex mercatoria could be equated with trade usage, there is still added problem that the content of the good faith obligation is difficult to ascertain. Take the supposed duty to renegotiate. Carbonneau points out that:

The concept of good faith is beset with so many problems that it threatens to be merely an empty label. It is thus submitted that it would be unsafe to find an international obligation of good faith based on Article 9 of the CISG. The issue of trade usages sharply divided the developing countries from the developed ones during the negotiations on the CISG.[154] There was a perception that many of the prevailing trade usages were created and influenced by the developed countries. It was felt that such trade usages were not necessarily fair to the developing countries. The developing countries would thus surely be alarmed if a general obligation of good faith were to be found based on problematic evidence of international practice. If an international trade usage of good faith were to be found based on such evidence, it would undermine the confidence of the developing countries in the CISG.

Article 9, then, is a provision that should not be readily invoked. Parties should be bound by a trade usage only if they expressly agree to it. They should be taken to have agreed to it only if sufficiently consistent practice can be convincingly demonstrated from a representative slate.

At best, a usage of good faith may exist which is peculiar to a particular region or a particular trade sector. Thus, it is possible that the parties from two or more continental countries or a continental country and the United States may be required to observe a practice of good faith. There should, however, be no international usage of good faith.

This of course does not preclude the possibility that specific trade usages may exist which prohibit one from engaging in certain behaviour on the ground that it would be tantamount to conducting oneself in bad faith. However, this is still very different from a trade usage which requires the general observance of good faith.

a. Good faith as a general principle

We now turn to the main arena of the debate - Article 7. The commentators who discuss the issue of good faith can broadly be divided into two camps: those who believe that good faith is a mere interpretative tool and those who believe that the doctrine of good faith has a more substantive role to play in the CISG. It is submitted that there are two main weaknesses with the debate thus far.

The first is a problem of coverage. Those who argue for an interpretative role for good faith base their argument largely on the wording of Article 7(1) of the Convention and the travaux préparatoires. They ignore the argument that good faith may well be one of the general principles on which the Convention is based. Those who tout a more substantive role for good faith, on the other hand, tend to base their arguments solely on Article 7(2) of the CISG and ignore or downplay the fact that that the Article 7(1) and the travaux préparatoires indicate otherwise.

The second is a problem of definition. Those who argue for an interpretative role fail to identify the definition of "interpretation" that they have in mind. In fact, some who claim to support merely an interpretative role for good faith appear to have a more substantive role in mind.[155] Similarly, those who argue for a more substantive role for good faith fail to identify the version they have in mind. For example, would good faith merely qualify what is expressly stated in the Convention? Or would it independently resolve matters not expressly settled by the Convention?

More fundamentally, both camps have a tendency to treat the meaning of "good faith" as self-evident. However, as it was pointed out above, "good faith" is susceptible to a variety of meanings.

It is hoped that this paper will avoid these pitfalls. The argument that good faith is one of the general principles on which the Convention is based will be addressed in this sub-section while the argument that good faith is a mere interpretative principle will be addressed in the next.

In the CISG, a general principle within the meaning of Article 7(2) can be found in two ways.[156] Firstly, an analogy can be drawn with provisions so closely related to the situation at hand that it must be conceded that the Convention could not have intended a different result. Secondly, one can seek a single overarching principle that undergirds the Convention as a whole.

It has also been argued that contracting parties may be bound by general principles not found within the Convention if it can be proved that they are "internationally coordinated and can actually find general acceptance".[157] In this way, the lex mercatoria or certain principles of the UNIDROIT Principles of International Commercial Contracts may come to bind contracting parties.[158]

It is submitted that this argument should be dismissed as out of hand. Quite apart from the difficulty of actually ascertaining principles that are "internationally coordinated and can actually find general acceptance", the suggestion directly contradicts the express direction in Article 7(1) that the general principle must be one on which the CISG is based. To insist otherwise is to rewrite the Convention.

Coming back to the concept of good faith, the most popular argument is that the concept underlies so many provisions of the CISG that it is a general principle that supports the entire Convention. As the Secretariat Commentary to the 1978 Draft of the Convention explained:

Those commentators who support a more substantive role for good faith base their arguments on a similar analysis.[160] An alternative argument, albeit less invoked, is that the general principle of good faith is one that is expressly mentioned in Article 7(1) of the Convention.[161] Thus, it is said that good faith may be used to resolve questions that concern matters governed by the Convention, but are not expressly settled by it. It is submitted, however, that there are three main problems with this theory:

i. There is no coherent content to the concept of good faith in the Convention.

The argument that good faith is a general principle of the CISG simply because it is expressly mentioned in Article 7(1) should be dismissed immediately. A concept does not become a general principle simply because it is expressly mentioned in one of the provisions of the Convention. If that were true, the Convention would be inundated with an overwhelming number of general principles. Rather, before good faith can qualify as one of the general principle of the CISG, it must be proved that it is a concept that runs through the entire Convention.

Unfortunately, in their eagerness to promote the idea that good faith is one of the general principles on which the Convention is based, commentators tend to latch onto whatever promotes justice or fairness in the Convention. The Secretariat Commentary was but one example. The following (sans footnotes) is another:

The problem with this approach is that it impliedly subscribes to such a general notion of good faith that it is practically useless. It is not concrete enough to generate answers to questions that have not been expressly settled by the CISG. Neither is it sufficiently nuanced to capture the subtle differences among the various rights and obligations that are said to be comprised in the concept of good faith. It has been pointed out that the provisions that are normally prayed in aid of a general principle of good faith can be parsed into at least three main so-called good faith duties: the duty to engage in good faith communication, the duty to refrain from conduct knowingly calculated to frustrate the contract and the duty to engage in good faith efforts to save the contract or mitigate damage resulting from breach.[163] Each duty has been created to meet different exigencies and therefore naturally generates different requirements. The danger of collapsing all these duties under the umbrella term of "good faith" is that it tends to erase the differences between them and makes them less sensitive to the context. While the CISG does permit the use of general principles to resolve matters not expressly settled by the Convention, one should be wary of inferring principles that are normatively pleasing but are at such a high level of abstraction that they are of limited utility.

Professor Schlechtriem is one commentator who has tried to place some coherent content to the concept of good faith in the Convention.[164] He argues that the standard of the "reasonable person" underlies so many provisions of the CISG that good faith must be one of the general principles on which the Convention is based.[165]

There are several objections, however, to his equation of "reasonableness" with "good faith". The first is that his argument is circular. Professor Schlechtriem manages to find a general principle of good faith in the CISG only because he starts out with the a priori assumption that the concept of "reasonableness" is synonymous with "good faith". Professor Schlechtriem's argument only works, however, if we can all agree that "reasonableness" is the same as "good faith".[166] Unfortunately, Professor Schlechtriem does not explain why he chooses "reasonableness" as the definition of good faith over the number of competing definitions that are available.

The second objection (and perhaps against any attempt to pin only one definition to such an elusive concept as "good faith") is that Professor Schlechtriem does not explain why "good faith" should be equated only with "reasonableness". The concept of good faith may well be capacious enough to embrace other ideas as well. Indeed, there are a number of other general principles in the Convention that also promote justice, fairness and ethical behaviour. The following are but four examples:[167]

All these general principles can be said to be manifestations or examples of a general idea of good faith and fair dealing. It is little wonder since much of the law is based on what is fair and what will achieve justice. The general principle of "reasonableness" now is thus but one particular manifestation of a still broader concept of good faith. Equating "reasonableness" with "good faith" has the unfortunate effect of implying that it is the sole means to achieve equity and threatens to overshadow all the other principles that work in tandem to achieve a satisfactory resolution of a case.

It is often not necessary, however, to think of these specific principles in terms of good faith since their applicability to a specific problem depends more on whether the factual matrix satisfies the requirements of the rule rather than whether the parties have satisfied some general idea of good faith. This approach is very similar to the piecemeal approach of the English common law. For that matter, it is also rather similar to the German approach. As we saw above, there is no one overarching definition of "good faith" in the German context. Instead, it is a label for a collection of specific rights and duties.

The irony of this objection is that a comprehensive substantive doctrine of good faith now becomes unnecessary and perhaps even undesirable. Any definition of good faith that strives to be comprehensive is likely to be too general. Any definition that attempts to be concrete is likely to be too narrow. It is thus far better to craft sensitive responses to particular problems through a piecemeal approach rather than resort to an amorphous doctrine of good faith with questionable content. One should thus treat good faith as a moral aspiration, but not as a substantive legal doctrine. At most, the term "good faith" serves as a useful catchall term for more particular principles.[168]

To reiterate, it is submitted that there is nothing wrong with subscribing to the very general notion that "good faith" is synonymous with anything that requires contracting parties to behave in a manner that ensures justice and fairness. The merit of this approach is that we do not omit anything that can assist us in circumscribing unethical behaviour. This idea, however, is not concrete enough to form the basis of a workable legal doctrine. Thus, we should concentrate on developing those principles in the CISG that manifest this general idea more specifically. Examples of such principles would be the general principle requiring the observance of reasonableness or the general principle requiring the mitigation of damage flowing from a breach of contract. The more general notion of good faith remains in the background as a moral aspiration or as a unifying thread to these more specific ideas. However, as discussed earlier, it is not absolutely crucial to think of these principles in terms of good faith.

ii. The imposition of substantive obligations of good faith on contracting parties would undermine the objective of the CISG to promote certainty and predictability in international trade.

Given the lack of coherence in the concept of good faith in the CISG, it would be foolhardy to promote it as one of the general principles of the CISG. It is significant that throughout the debate on the appropriateness of inserting an obligation of good faith in the Convention, none of the delegates actually explained precisely what an obligation of good faith in the international context would entail.

It is too optimistic to expect national courts to be able to explicate on the content of an obligation of good faith in international transactions when domestic jurisdictions are still wrestling with the appropriate meaning of good faith in the domestic context. Decision-makers would inevitably draw on their varied experience with the doctrine in the domestic context, thereby producing multiple contradictory answers. While some semblance of an international doctrine may finally emerge, this would be an extremely long time in coming and would undermine the certainty, predictability and stability of international transactions in the meantime.

Since the objectives of certainty, predictability and uniformity are paramount in the CISG, it is best to use the concept of good faith in a way that will do the least damage. While the absence of a general principle of good faith may mean that decision-makers would have to turn to the rules of private international law to resolve an issue in so far as some other general principle in the CISG cannot be found to assist, this would still be preferable since it is likely that the domestic law would already have developed a response to the problem. In this way, domestic notions of good faith may be used to resolve a problem. Although this may produce different solutions, it is far better than having to contend with the "wild-card" of an uncertain general principle of good faith.

While it is recognised that it would be highly desirable to promote good faith and fair dealing in international trade, the adoption of a "loose cannon" is not the best way to achieve it. It is best not to force the issue before the doctrine of good faith is fully developed (if ever) in international trade law.

iii. The negotiating history of the CISG reveals that Article 7 is a clear compromise between those who wanted the Convention to impose a substantive obligation of good faith on contracting parties and those who were against a reference to good faith.

The third and final problem is that the thesis that good faith is one of the general principles of the CISG blatantly contradicts the negotiating history of Article 7. As Professor Farnsworth puts it, Article 7(1) is a "statesmanlike compromise".[169] It skilfully strikes a middle ground between the civilian and common law traditions. Imposing a full-blown doctrine of substantive good faith obligations would have been too much and too novel for many common law jurisdictions. On the other hand, representatives from civilian jurisdictions would not have been appeased without some reference to good faith in the Convention.[170]

It would pervert the compromise to now use Article 7(2) to impose substantive obligations of good faith and fair dealing on the parties.[171]

If the drafting committee really did intend to impose substantive obligations on the parties, it could well have adopted the much more straightforward wording of Article 1.7 of the UNIDROIT Principles of International Commercial Contracts which reads:

The reason why the UNIDROIT Principles could go where the CISG could not was because:

Of course, this is not to say that all the choices UNIDROIT ultimately made were necessarily the most appropriate for international trade. However, at the very least, UNIDROIT's determinations were not obfuscated by the "confounding language of diplomatic squabbling".[173]

b. Good faith as an interpretative tool

The only role that good faith as a legal doctrine could now possibly play in the CISG would be as an aid to interpretation. This would be consistent with the literal wording of Article 7(1), the negotiating history of the Convention and most importantly of all, would avoid all the problems associated with a more substantive version of good faith.

Before the feasibility of this interpretation is further discussed, one contrary argument must first be addressed. Some commentators who support the imposition of a substantive good faith obligation have managed to enlist even the wording of Article 7(1) in support. Their argument is that the terms of the contract impose a general obligation of good faith because there can be no distinction between the interpretation of the CISG and the interpretation of the contract itself.[174]

While it is true that the interpretation of the CISG does impose obligations and confer rights on contracting parties just as a contract does and that the contract itself derives its binding force from the Convention, there is still no reason to ignore the distinction between the two. It is still one thing to determine the meaning of the CISG by interpreting its provisions and yet another to determine the meaning of the contract by interpreting its terms. It is the CISG's provisions as interpreted that impose obligations and confer rights on parties. While the provisions of the CISG may sometimes aid in the interpretation of the contract, it is nevertheless its provisions as interpreted that assist. If the argument of these commentators is accepted, it would lead to the surprising result that there is a general obligation of good faith in all possible cases - a result that was expressly rejected during the negotiations on the CISG.[175]

With this argument out of the way, it must still be determined what the words "interpret" and "good faith" in the context of Article 7(1) mean. In particular, it must be determined if it is even possible for "good faith" to have any coherent meaning here when it had been so difficult to define it for more substantive purposes.

It is submitted that the most natural and ordinary meaning of "interpret" should be adopted i.e. "to explain or tell the meaning of" or "to make understandable".[176] Thus, regard should be had to the need to promote the observance of good faith in international trade in order to resolve cases of textual ambiguity. The meaning of a provision can be ambiguous either because it naturally bears a number of multiple meanings or because it must be applied to a previously unforeseen situation.

Honnold is one commentator who formally accepts that the role of good faith is restricted to one of interpretation.[177] He argues, however, that the concept of good faith should be used, inter alia, to deny either the buyer or the seller the remedies of specific performance or avoidance when either party seeks to use these remedies to speculate at the other's expense.[178] In other words, good faith is a prerequisite to the exercise of any of the rights or remedies in the CISG. It is submitted that this crosses the line of legitimate interpretation to the imposition of substantive obligations of good faith on the parties. This method of "interpretation" does not merely seek to pin down one meaning among the multiple interpretations that the express wording of the provision can fairly bear. It is tantamount to "gap-filling" or implying qualifications into the CISG which the instrument has not expressly provided for.

We now turn to the more thorny question of what "good faith" in the context of Article 7(1) means. We already saw how impossible and perhaps even undesirable it was to define the concept in the context of Article 7(2) of the CISG. Perhaps in that light, the CISG should not have made any reference to the concept of good faith at all. While Article 7(1) may have been an adroit compromise in political terms, its merit in substantive legal terms is dubious.

Nevertheless, the term "good faith" is used in the context of Article 7(1). If the concept is to prove to be of any assistance in the interpretation of the Convention, some coherent meaning must somehow be found. It is submitted that it is not contradictory to attempt to prescribe a meaning of good faith for the purposes of Article 7(1) while refusing to do so for Article 7(2). There were good reasons to resist doing so for the purposes of Article 7(2) because both policy reasons and the Convention's negotiating history were against it. Good faith in the form of an interpretative principle, on the other hand, is far less threatening to the objectives of the CISG. Moreover, the Convention has given its imprimatur to the task by specifically referring to the concept in Article 7(1).

Unfortunately, the travaux préparatoires do not provide any guidance beyond the fact that the concept is not intended to play a robust substantive role. The academic literature on the CISG is also unhelpful on this point since it generally either assumes that the definition of good faith is self-evident or that the concept will somehow magically be elucidated by case law.[179]

There are, however, several guidelines that may be observed in the attempt to ascribe some meaning to the term.

• Firstly, we have more flexibility in defining "good faith" in the context of Article 7(1) than we had in Article 7(2). This is because in so far as Article 7(2) of the CISG was concerned, good faith would qualify as one of the general principles of the Convention only in so far as it could be shown that it pervaded the entire Convention. It followed, therefore, that the meaning of the principle, if it existed, had to be derived from the Convention itself. No external sources could be used. Any definition had to be expansive enough to encompass all the concepts or principles within the Convention that were manifestations of a more general idea of good faith. However, as discussed above, no coherent meaning of good faith could be derived from the Convention itself. There were so many principles affiliated to the concept of good faith in the CISG that one could not achieve a comprehensive definition which still retained concrete content.

  Where good faith in the context of Article 7(1) is concerned, however, we are not similarly constrained. The meaning of good faith need not come from within the Convention but can also be found externally. Therefore, the most appropriate meaning can be chosen from the competing definitions that are available.

• Secondly, we can afford not to be overly ambitious in choosing a definition. If one has to choose between a comprehensive but general definition and a narrow but concrete one, the choice should be in favour of the latter. It is best to choose a limited definition that everyone, at a minimum, would agree on. While this may lay the definition open to a charge of under-inclusiveness, it would minimise the disruption that an uncertain concept of good faith could wreak on the structure of the CISG and promote the cause of predictability and stability. If one had to define good faith for the purposes of Article 7(2), on the other hand, one would have to engage in over-abstraction due to the number of concepts that were affiliated to the idea of good faith.

• Thirdly, to state the obvious, the definition of good faith must be one that would aid in the interpretation of the Convention. It would be useless to choose a standard that would be helpful in generating substantive obligations but not in resolving interpretative problems.

Unfortunately, these guidelines do not advance the inquiry very much. The definition of "good faith" remains as elusive as ever. We still face many of the same problems that we encountered in trying to define "good faith" for the purposes of Article 7(2). There are still so many competing definitions to contend with that it is difficult to achieve a comprehensive definition that is capable of practical guidance. The problem is exacerbated by the fact that we now have to contend not only with the specific principles of good faith in the Convention, but also with a number of definitions that are externally derived as well. Thus, the danger is that any definition would "either spiral into the Charybdis of vacuous generality or collide with the Scylla of restrictive specificity".[180] It would not be appropriate to simply choose one definition over another since that would mean sacrificing nuances captured by one term but not another.

Neither is it possible to ascertain some "minimum agreed standard" of good faith. As discussed above, almost every jurisdiction has a very different conception of what it means to act in good faith. It would be too much to hope that they would be able to agree on a minimum standard for the purposes of international trade.

Ironically, perhaps, we are thus forced to fall back on the very same conception of good faith that we used for the purposes of Article 7(2). To repeat, the conclusion there was that the CISG does not embody a general principle of good faith, at least not in a form that it is capable of independently generating concrete rights and obligations. It was argued, however, that the Convention does contain several principles that can be said to be particular manifestations of a broad idea of good faith. Examples of such principles were the principles of reasonableness, estoppel and favor contractus.[181]

It is submitted that this conception of good faith can also be utilised for the purposes of Article 7(1). The term "good faith" in Article 7(1) should not be read to refer to some general doctrine of good faith, for no coherent one can be formulated. Instead, it should be read as a compendious term referring to the collection of more specific good faith principles in the Convention. Thus, when a question of interpretation arises, the tribunal should first decide which of these principles are implicated and resolve the problem in a way that upholds that principle.[182] Thus, the definition of "good faith" for the purposes of Article 7(1) is one that is derived from the CISG itself. It is true that there are a number of other good faith definitions that can be externally derived, either from national jurisprudence or academic literature on the subject. It is submitted, however, that these definitions are not as strong as one that is internally derived from the CISG. This is because there is no guarantee that any of these "external" definitions will meet with the approval of those countries that have adopted (or will adopt) the Convention. At a minimum, however, they must accept those principles embodying the idea of good faith because they are contained in the Convention itself.

The upshot of all this is that the one reference that the CISG does make to the concept of good faith should not be understood as sanctioning the adoption of a legal doctrine of good faith, even if it is only for interpretative purposes.

The following example will illustrate how the above points would apply. A provision that has taken on an especially ambiguous meaning in this technological age is Article 18(2) which provides: "An acceptance of an offer becomes effective at the moment the indication of assent reaches the offeror [emphasis added]."

In so far as an acceptance by electronic mail is concerned, an acceptance may be deemed to have "reached" the offeror at any one of three points- when it is stored in the computer, when it is read or when it is reproduced in a tangible form.[183] The matter is further complicated if the offeror fails to check his electronic mail diligently and only discovers the message after the date for acceptance has passed or intentionally waits until the time has passed for acceptance before reading his e-mail.

The concept of good faith as explained above can be used here to identify the appropriate definition of "reaches". The specific good faith principle that is implicated here is the one that requires the observance of reasonable behaviour in the absence of any other specific regulation. The message should thus not be deemed to have "reached" the offeror when the message first arrives and is stored on the computer's hard disk. It would be unreasonable for the offeree to expect that that the offeror has knowledge of the acceptance even before he has had a chance to read it. It is only fair that the offeror should be given a reasonable opportunity to read the message. Thus, the acceptance should be considered to have "reached" the offeror only after he has had a reasonable opportunity to read the message. What qualifies as a "reasonable opportunity" would depend on factors such as the offeror's known access to an e-mail server and whether the offeree sent the message outside office hours.[184] It would constitute bad faith,[185] however, for the offeror to insist that an acceptance "reaches" him only after he has printed it because he is, by then, already cognisant of the acceptance. If it were otherwise, the offeror would be free to act in bad faith and print the message only after the time for acceptance has passed, thus thwarting the reasonable expectations of the offeree.

To summarise the conclusions of this part of the paper, a general doctrine of good faith in a form that is capable of generating independent legal rights and obligations for either substantive or interpretative purposes does not exist in the CISG. At best, general principles may be derived which are particular manifestations of a very general idea of good faith. Where possible, these general principles can be used to resolve matters not expressly settled by the Convention. These same principles can also be used to resolve issues of textual ambiguity.

It is acknowledged that some may object to the equation of "good faith" with anything that promotes justice and fairness or moderates unethical behaviour on the ground that it merely replaces one vague and nebulous term with equally vague and nebulous synonyms.

It is submitted, however, that the problem is not as grave as it seems. This is because it is not the thesis of the paper that the concepts of "justice", "fairness" and "ethics" should be left to generate rights and obligations on their own. They are merely convenient labels for some of the general principles that are embodied within the Convention. It is actually these principles that should be used to resolve issues, not the concepts themselves. Thus it does not really matter even if we were to disagree over whether a principle does indeed promote justice and fairness or moderates unethical behaviour. It remains a fact that it nevertheless is still one of the general principles on which the CISG is based. Thus, regardless of the nomenclature used, it still can be used to resolve matters governed by the Convention but not expressly settled in it.

The only area where a disagreement could conceivably make a difference is when we seek to use these same principles to resolve issues of textual ambiguity. In this context, nomenclature does matter. This is because Article 7(1) refers specifically to the concept of good faith. Thus, one can only make use of those principles that manifest the idea of good faith to interpret the Convention. However, decision-makers may well use different principles to resolve issues of textual ambiguity if their ideas of justice, fairness and ethics are different.

While this is indeed a problem, its magnitude should not be exaggerated. As long as the reference to good faith in the Convention remains, the problem of uncertainty and vagueness can never be completely eliminated. The only thing we can do is to minimise the problem. Restricting the definition of "good faith" in Article 7(1) to encompass only those principles in the CISG that constrain the behaviour of contracting parties in order to promote justice, fairness and ethical behaviour achieves this in three ways.

Firstly, it severely circumscribes the scope of decision-makers' discretion by limiting the menu of options they can choose from. Instead of giving them the option to resort to a free-ranging principle of good faith, decision-makers now are limited to those general principles of the CISG that manifest this idea of good faith. This is far better than leaving a loose cannon in their hands. In order to ensure that this version of good faith actually does operate as a significant constraint, decision-makers should be expected to demonstrate that the principle that they seek to use is indeed one on which the CISG is based and to articulate precisely how that principle promotes justice and fairness, or moderates unethical behaviour.

Secondly, even if decision-makers were to use a general principle that does not promote the cause of good faith to resolve an issue of interpretation, the damage done is not acute. This is because the role of good faith is restricted to issues of textual interpretation only. Thus its impact cannot be as severe as a substantive, wide-ranging principle.

Thirdly, even if the principle used does not embody some specific idea of good faith, we can at least seek solace from the fact that the errant decision-maker must be promoting some other value of the Convention since he is employing one of the principles on which the CISG is based.

IV. THE APPLICATION AND ENFORCEMENT OF ARTICLE 7

A. A Confusion of Judicial and Arbitral Determinations

This part of the paper now turns to examine whether judicial and abritral tribunals have managed to capture the correct definition of good faith in their rulings.

A survey of the field,[186] however, reveals that judicial and arbitral determinations have only exacerbated the confusion over the proper role of good faith in the Convention. It is possible to find a decision that supports almost every possible interpretation of the role of good faith in the CISG. At times, these courts or tribunals issue their decisions without any recognition of the debate surrounding the concept of good faith in the Convention and without an examination of the ramifications of its findings on the role of good faith in the CISG.

To illustrate this point, I will follow the six-way division I adopted in Part III (B)(1) ("The Competing Interpretations of the Role of Good Faith in the Convention") of this paper and highlight a decision that best exemplifies each interpretation.

1. An obligation of good faith established by the practices of parties or by international trade usage.

No cases were found that support this interpretation.

2. The doctrine of good faith as an interpretative guide.

The ICC arbitration case of W v. R [187] involved a German seller and a Spanish buyer who had concluded an agreement for the buyer to be the exclusive distributor in Spain of industrial equipment produced in Germany. Several individual sales contracts were concluded pursuant to this agreement. Four years later, the German seller terminated the exclusive distribution agreement due to insufficient sales. The arbitration involved several claims but the one that is relevant for our purposes is the buyer's counterclaim for damages arising from the seller's failure to deliver spare parts. It was held that the CISG was not applicable to the exclusive distribution agreement but to the individual sales agreements.

The arbitrator recognised that the seller would have had an obligation to deliver spare parts if the case were based on German law. Under Article 433 of the BGB, the producer of series-produced automobiles, machines and technical equipment has a collateral obligation to have replacement parts ready for delivery for a certain period of time, even if there has been no special agreement. This same result could also be arrived at through § 242 of the BGB or Article 26(2) of the Law Governing Competition Limitation.

The arbitrator acknowledged that there was a debate over the role of good faith in the Convention. He maintained, however, that a collateral obligation similar to that imposed by Germany could not be inferred from Article 7(1) of the CISG because it was relevant only to the interpretation of the Convention. Unfortunately, for our purposes, the arbitrator did not go on to elaborate on the definition of good faith he would use to interpret the Convention nor the version of interpretation that he had in mind.

It must nevertheless be noted that the absence of a substantive good faith obligation in the CISG did not prevent the arbitrator from arriving at an equitable result. He found that "a prompt delivery of replacement spare parts had become normal practice as defined by Article 9(1) of the CISG".[188] In accordance with Article 33(c) read with Article 7(2) of the CISG, the seller was obliged to deliver the spare parts within a reasonable time.

3. The doctrine of good faith as a prerequisite to the exercise of the rights and remedies provided in the Convention.

A 1995 case heard before the OLG München supports this interpretation.[189] In this case, the plaintiff buyer had concluded a contract with the defendant seller for eleven cars. Five of these cars were ready for delivery in August and the remaining six in October. The buyer, however, was unable to take delivery of the cars due to extreme currency fluctuations and asked the seller to defer their delivery until the situation returned to normal.

The issue that is relevant for our purposes is the buyer's claim two and a half years later for damages for non-delivery. This claim was based on Articles 45(1)(b), 45(2), 49(1)(a) and 25 of the CISG. The buyer's argument was that the seller's failure to deliver the cars when it wished constituted a fundamental breach of contract. The court held, however, that as the parties had not agreed on a precise date of delivery, the seller's readiness to deliver in August and October meant that it was not guilty of non-delivery.

In any case, allowing the buyer to avoid the contract now, two and a half years after the event, would violate the principle of good faith in Article 7(1) of the CISG.[190]

Unfortunately, the court did not elaborate further on its interpretation of the CISG. One commentator speculates that "the court may have interpreted the provisions relating to avoidance, particularly article 49 which gives the parties the power to avoid, and article 26 which provides that avoidance is only effective if notice is given, with a view to promoting good faith in international trade as directed in article 7(1)".[191] One should note, however, that nowhere in Article 49 and Article 26 is there a requirement that the buyer must make his claim in a timely fashion before he can avoid the contract in cases where the seller has not delivered the goods. The court's holding, therefore, must have been predicated on an unspoken assumption that all parties had to observe good faith before they could invoke the rights and remedies of the Convention. This amounts to the imposition of substantive obligations of good faith on the party and contradicts the express wording of Article 7(1).

It is a pity that the court felt that it had to resort to the "blunt cleaver" of a good faith obligation when it could have availed itself of more specific principles that would have been more sensitive to the context. It should have noted that in cases where the seller has delivered the goods, Article 49(2) deprives the buyer of his right to avoid the contract if he fails to do so within a reasonable time and in any event, after he knew or ought to have known of the breach. In cases where the seller has delivered non-conforming goods or goods that are subject to third party claims, Article 39 and 43 require the buyer to exercise his remedies a reasonable time after he knew or ought to have been aware of the breach. Evidently, the CISG places a clear premium on the timely exercise of one's rights and remedies. The court could have drawn an analogy between these provisions and the present case and concluded that the CISG could not h